Aqueous Lithium–Iodine Solar Flow Battery for the Simultaneous

Jun 23, 2015 - Integrating both photoelectric-conversion and energy-storage functions into one device allows for the more efficient solar energy usage...
1 downloads 9 Views 776KB Size
Communication pubs.acs.org/JACS

Aqueous Lithium−Iodine Solar Flow Battery for the Simultaneous Conversion and Storage of Solar Energy Mingzhe Yu, William D. McCulloch, Damian R. Beauchamp, Zhongjie Huang, Xiaodi Ren, and Yiying Wu* Department of Chemistry & Biochemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210, United States S Supporting Information *

energy. Using solar energy enables the Li−I SFB to be charged at an input voltage of 2.90 V under 1 sun 1.5 AM illumination and discharged at an output voltage of 3.30 V (at the current density of 0.50 mA cm−2). Compared to conventional Li−I batteries, which are typically charged at a voltage over 3.60 V, the Li−I SFB can achieve energy savings up to 20% because of its voltage reduction. Furthermore, we have also proved with our example of the sodium−iodine (Na−I) SFB that this SFB concept can be extended to other metal−redox solar flow batteries. The Li−I SFB has a three-electrode configuration (Figure 1a): a metallic Li anode, a Pt counter electrode (CE) and a dyesensitized TiO2 photoelectrode (PE). Both the CE and PE are in contact with the flowing I3−/I− redox catholyte, which is stored in a reservoir connected to the catholyte chamber and pumped through the device using a peristaltic pump. The Li anode and I3−/I− catholyte are separated by a piece of ceramic Li-ion

ABSTRACT: Integrating both photoelectric-conversion and energy-storage functions into one device allows for the more efficient solar energy usage. Here we demonstrate the concept of an aqueous lithium−iodine (Li−I) solar flow battery (SFB) by incorporation of a built-in dyesensitized TiO2 photoelectrode in a Li−I redox flow battery via linkage of an I3−/I− based catholyte, for the simultaneous conversion and storage of solar energy. During the photoassisted charging process, I− ions are photoelectrochemically oxidized to I3−, harvesting solar energy and storing it as chemical energy. The Li−I SFB can be charged at a voltage of 2.90 V under 1 sun AM 1.5 illumination, which is lower than its discharging voltage of 3.30 V. The charging voltage reduction translates to energy savings of close to 20% compared to conventional Li−I batteries. This concept also serves as a guiding design that can be extended to other metal-redox flow battery systems.

S

imultaneous conversion and storage of solar energy marks a significant advance toward practical solar energy usage. Various kinds of solar fuels have been actively explored.1 However, challenges with hydrogen storage and the cost of fuel cells make those systems complicated and difficult for implementation. A promising solution is integrating a photoelectrode into an electrochemical capacitor or battery to form a single device.2 Several groups have made pioneering contributions toward this goal. For instance, the integration of dyesensitized photoelectrochemical cells with redox flow batteries has been explored.3 Our group has also demonstrated photoassisted charging of a Li−O2 battery.2d However, these devices are limited because they use organic solvents for the electrolyte. Besides the cost and the negative environmental impact of organic solvents, these systems cannot be incorporated with current aqueous redox flow battery systems because of their incompatible organic-solvent design. Recently, attempts at creating aqueous systems have been made by replacing the dye-sensitized photoelectrode with semiconductor photoelectrodes that have hydrophilic surfaces.4 However, due to the semiconductors’ large band-gap (i.e., 2.7 eV for WO3 and 3.2 eV for TiO2), these devices can only harvest a very limited portion of the solar spectrum (i.e., 3 V) but also is more environmentally friendly and cost-effective. Furthermore, this work’s concept of combining battery electrochemistry with solar cell photoelectrochemistry also serves as a guiding design that can be extended to other metal−redox flow battery systems.

indicating that a photovoltage, which compensates for the cell’s charging voltage, is generated on the PE. Once the light is turned off, the charging voltage increases instantly. The light response can also be seen in the Nyquist plots (Figure 4b), which were obtained through electrochemical impedance spectroscopy (EIS). The semicircle at low frequency is attributed to the recombination charge-transfer process at the PE−catholyte interface. Under illumination, the high electron concentration in the TiO2 conduction band and sub-bandgap trap states induces a much smaller charge-transfer resistance. Because the Ef drops in the dark, the TiO2 becomes more insulating and the recombination mainly occurs through the FTO−catholyte interface.11 The comparison between the charging profile of a conventional Li−I battery and the Li−I SFB is shown in Figure 5. Under

Figure 5. Performance comparison between Li−I SFB and conventional Li−I batteries: (a) a typical depleted charging and discharging voltage profile; (b) the initial charging voltages for 25 cycles.

1 sun 1.5 AM illumination, the Li−I SFB has an initial charging voltage of 2.90 ± 0.01 V at a current density of 0.50 mA cm−2 based on the average of three devices (Figure 5a). This value matches well with theoretical predictions, considering the energy gap between the TiO2 CBM (i.e., +2.7 V at pH = 4.6, the pH of aqueous catholyte) and the Ef of electrons at working condition as well as the electrolyte-dye recombination and the device’s internal series resistance. Compared to the conventional Li−I battery, which has initial charging voltage of 3.60 V, the Li−I SFB achieves a voltage reduction of 0.70 V (3.60 − 2.90 = 0.70 V), which translates to energy savings of close to 20% (0.70 V/3.60 V × 100% = 19%). This charging voltage is even lower than the discharging voltage, which is thermodynamically impossible without the solar energy input. As the charging process proceeds, the accumulation of I3− and consumption of I− shifts the I3−/I− redox potential positively and reduces the catholyte’s ionic conductivity. Therefore, the charging voltages of both devices gradually increase due to the increasing internal resistance and the more-severe electrolyte− dye recombination (for the Li−I SFB case). At a cutoff voltage of 3.6 V, the solar battery with 0.100 mL of catholyte (2.00 M LiI, 0.50 M GuSCN in saturated Cheno aqueous solution) is able to be photocharged to a volumetric capacity of 32.6 Ah L−1 in 16.80 h, which is 91% of its theoretical capacity (i.e., 35.7 Ah L−1 for the catholyte with 2.00 M LiI). This value is close to the capacity of conventional Li−I batteries in the literature.5b Our Li−I SFB also demonstrates good cyclability. As shown in Figure 5b, the initial charging voltage remains stable at 2.91 ± 0.02 V for at least 25 cycles through continuous cycling. (See SI for experimental details and capacity calculations of Figure 5a,b.) Although the volumetric capacity of 35.7 Ah L−1 is promising for practical applications, it should be noted that the current system is limited by the low photocharging rate (i.e., 16.80 h per 0.1 mL



ASSOCIATED CONTENT

S Supporting Information *

Experimental details, discussion on CV of aqueous and aprotic catholyte based Li−I SFB, molecular structure of Z907 dye, results of DMSO-based Li−I SFB, and results of Na−I SFB. The 8334

DOI: 10.1021/jacs.5b03626 J. Am. Chem. Soc. 2015, 137, 8332−8335

Communication

Journal of the American Chemical Society

(13) (a) Lu, Y.; Goodenough, J. B.; Kim, Y. J. Am. Chem. Soc. 2011, 133, 5756−5759,. (b) Wang, Y.; Wang, Y.; Zhou, H. ChemSusChem 2011, 4, 1087−1090,. (c) Zhao, Y.; Ding, Y.; Song, J.; Li, G.; Dong, G.; Goodenough, J. B.; Yu, G. Angew. Chem., Int. Ed. 2014, 53, 11036− 11040,. (d) Zhao, Y.; Ding, Y.; Song, J.; Peng, L.; Goodenough, J. B.; Yu, G. Energy Environ. Sci. 2014, 7, 1990−1995,. (e) Lu, Y.; Goodenough, J. B. J. Mater. Chem. 2011, 21, 10113−10117,. (f) Li, N.; Weng, Z.; Wang, Y.; Li, F.; Cheng, H.-M.; Zhou, H. Energy Environ. Sci. 2014, 7, 3307− 3312,.

Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.5b03626.



AUTHOR INFORMATION

Corresponding Author

*[email protected] Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We acknowledge funding support from the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Science and Engineering (Award: DE-FG02-07ER46427). We thank B. B. Trang for her detailed and valuable comments on the manuscript.



REFERENCES

(1) (a) Gray, H. B. Nat. Chem. 2009, 1, 7. (b) Magnuson, A.; Anderlund, M.; Johansson, O.; Lindblad, P.; Lomoth, R.; Polivka, T.; Ott, S.; Stensjö, K.; Styring, S.; Sundström, V. Acc. Chem. Res. 2009, 42, 1899−1909,. (c) Schlapbach, L.; Zuttel, A. Nature 2001, 414, 353−358,. (d) Alibabaei, L.; Luo, H. L.; House, R. L.; Hoertz, P. G.; Lopez, R.; Meyer, T. J. J. Mater. Chem. A 2013, 1, 4133−4145,. (e) Khaselev, O.; Turner, J. A. Science 1998, 280, 425−427,. (f) Maeda, K.; Teramura, K.; Lu, D.; Takata, T.; Saito, N.; Inoue, Y.; Domen, K. Nature 2006, 440, 295−295,. (2) (a) Chen, T.; Qiu, L.; Yang, Z.; Cai, Z.; Ren, J.; Li, H.; Lin, H.; Sun, X.; Peng, H. Angew. Chem., Int. Ed. 2012, 51, 11977−11980,. (b) Guo, W.; Xue, X.; Wang, S.; Lin, C.; Wang, Z. L. Nano Lett. 2012, 12, 2520− 2523,. (c) Fu, Y.; Wu, H.; Ye, S.; Cai, X.; Yu, X.; Hou, S.; Kafafy, H.; Zou, D. Energy Environ. Sci. 2013, 6, 805−812,. (d) Yu, M.; Ren, X.; Ma, L.; Wu, Y. Nat. Commun. 2014, 5, 5111. (e) Hauch, A.; Georg, A.; Krašovec, U. O.; Orel, B. J. Electrochem. Soc. 2002, 149, A1208−A1211,. (f) Nagai, H.; Segawa, H. Chem. Commun. 2004, 8, 974−975,. (g) Murakami, T. N.; Kawashima, N.; Miyasaka, T. Chem. Commun. 2005, 26, 3346− 3348,. (h) Chen, H.-W.; Hsu, C.-Y.; Chen, J.-G.; Lee, K.-M.; Wang, C.C.; Huang, K.-C.; Ho, K.-C. J. Power Sources 2010, 195, 6225−6231,. (i) Saito, Y.; Uchida, S.; Kubo, T.; Segawa, H. Thin Solid Films 2010, 518, 3033−3036,. (j) Wee, G.; Salim, T.; Lam, Y. M.; Mhaisalkar, S. G.; Srinivasan, M. Energy Environ. Sci. 2011, 4, 413−416,. (3) (a) Yan, N. F.; Li, G. R.; Gao, X. P. J. Mater. Chem. A 2013, 1, 7012−7015,. (b) Yan, N. F.; Li, G. R.; Gao, X. P. J. Electrochem. Soc. 2014, 161, A736−A741,. (c) Liu, P.; Cao, Y.-l.; Li, G.-R.; Gao, X.-P.; Ai, X.-P.; Yang, H.-X. ChemSusChem 2013, 6, 802−806,. (4) (a) Wei, Z.; Liu, D.; Hsu, C.; Liu, F. Electrochem. Commun. 2014, 45, 79−82,. (b) Liu, D.; Wei, Z.; Hsu, C.-j.; Shen, Y.; Liu, F. Electrochim. Acta 2014, 136, 435−441,. (c) Liu, D.; Zi, W.; Sajjad, S. D.; Hsu, C.; Shen, Y.; Wei, M.; Liu, F. ACS Catal. 2015, 5, 2632−2639,. (5) (a) Zhao, Y.; Byon, H. R. Adv. Energy Mater. 2013, 3, 1630−1635,. (b) Zhao, Y.; Wang, L.; Byon, H. R. Nat. Commun. 2013, 4, 1896. (c) Zhao, Y.; Hong, M.; Bonnet Mercier, N.; Yu, G.; Choi, H. C.; Byon, H. R. Nano Lett. 2014, 14, 1085−1092,. (6) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737−740,. (7) (a) Law, C.; Pathirana, S. C.; Li, X.; Anderson, A. Y.; Barnes, P. R. F.; Listorti, A.; Ghaddar, T. H.; O′Regan, B. C. Adv. Mater. 2010, 22, 4505−4509,. (b) Law, C.; Moudam, O.; Villarroya-Lidon, S.; O’Regan, B. J. Mater. Chem. 2012, 22, 23387−23394,. (8) Dean, J. L. Lange’s Handbook of Chemistry, 14th ed.; McGraw Hill: New York, 1992. (9) Wang, Y.; He, P.; Zhou, H. Adv.Energy Mater. 2012, 2, 770−779,. (10) Boschloo, G.; Hagfeldt, A. Acc. Chem. Res. 2009, 42, 1819−1826,. (11) Wang, Q.; Ito, S.; Gratzel, M.; Fabregat-Santiago, F.; Mora-Sero, I.; Bisquert, J.; Bessho, T.; Imai, H. J. Phys. Chem. B 2006, 110, 25210− 25221,. (12) (a) Kamat, P. V.; Tvrdy, K.; Baker, D. R.; Radich, J. G. Chem. Rev. 2010, 110, 6664−6688,. (b) Gratzel, M. Nature 2001, 414, 338−344,. 8335

DOI: 10.1021/jacs.5b03626 J. Am. Chem. Soc. 2015, 137, 8332−8335